Three-stage contingent screening for Down syndrome.

OBJECTIVE: To demonstrate the potential value of three-stage sequential screening for Down syndrome.
METHODS: Protocols were considered in which maternal serum pregnancy associated plasma protein-A (PAPP-A) and free beta-human chorionic gonadotropin (hCG) measurements were taken on all women in the first trimester. Those women with very low Down syndrome risks were screened negative at that stage and nuchal translucency (NT) was measured on the remainder and the risk reassessed. Those with very low risk were then screened negative and those with very high risk were offered early diagnostic testing. Those with intermediate risks received second-trimester maternal serum alpha-fetoprotein, free beta-hCG, unconjugated estriol and inhibin-A. Risk was then reassessed and those with high risk were offered diagnosis. Detection rates and false-positive rates were estimated by multivariate Gaussian modelling using Monte-Carlo simulation.
RESULTS: The modelling suggests that, with full adherence to a three-stage policy, overall detection rates of nearly 90% and false-positive rates below 2.0% can be achieved. Approximately two-thirds of pregnancies are screened on the basis of first-trimester biochemistry alone, five out of six women complete their screening in the first trimester, and the first-trimester detection rate is over 60%.
CONCLUSION: Three-stage contingent sequential screening is potentially highly effective for Down syndrome screening. The acceptability of this protocol and its performance in practice, should be tested in prospective studies. Copyright 2006 John Wiley & Sons, Ltd.